Black Anodized parts turning brown

A. Hi Brian. When you look at buildings with aluminum awnings, handrails, and roofs, you'll see that they are almost always within the color range from champagne to bronze. The main reason is that it is possible to deposit metal into an anodized coating which will generate this color and be completely resistant to UV fading, whereas other colors (including black) require organic dyes (similar to fabric dyes) which to varying degrees will suffer from UV bleaching.

We have appended your inquiry to one of several similar, long, and detailed threads. Some of it is highly technical and addressed to anodizing shops rather than end customers, but can be summarized as good black color stability requires thick anodizing, high quality UV-rated dyes, and proper sealing (here we refer to a process done in the shop to swell the anodize pores closed, not to any sort of clear coat).

I hate to be the bearer of bad news, but the parts which are badly fading are probably simply not of high enough quality for a motorcycle exposed to sun and heat.

Krylon seems to me to claim that their UV-resistant clear coat can protect stuff like this. I'm not so sure, but if you try it please let us know what happened.

Luck & Regards,

A. 1. The first guess is a seal problem. So you changed seals. Did you perform any seal testing to ascertain the sealing quality? Even Q-tip [affil links] dipped in acetone [on eBay or Amazon], if it picks up dye - it ain't sealed ! There are several ASTM seal tests.

2. Do you use an iron based deox? Is the iron dragged into the anodizing and occluding. Three of the iron oxides are "brown".

3.What is the dissolved aluminum in the anodizing tank? Too much clogs the pores and interferes with dye absorption and subsequent hydration of the aluminum oxide (sealing).

4. What is the anodizing temperature. It affects pore size which affects dying and sealing.

5. Anodize thickness affects dye absorption and sealing.

6. Has the customer changed aluminum alloy?

7. Light Fastness of the dye ?

A. If the parts leave your shop black, and then sometime later turn brown (less than black), the answer is one of three things:

1) The part is exposed to excessive Heat.

2) Lightfast quality of your dye stuff is poor. You could have a great seal, but if your dye stuff has poor lightfastness, your part will fade - when exposed to UV/sun.

3) Poor seal.

A. Dear sir,

Aluminum extrusion can be blackened by two step color anodizing provided the thickness of oxide layer is more than 10 microns.

A. Extrusions can be anodized and dyed black successfully. The color fading depends on the environment and conditions the parts are exposed to. Sunlight and Fluorescent light will fade the dye. If you achieve .70 mils (7 tenths of a thousandth) and are using a dye that has a good colorfast rating, and go for an architectural seal time in nickel acetate seal you will have the best results. Extrusions will also look much better if they are deoxidized prior to etching.

Good luck!

P.S Anodizing is not impervious to scratches, but it should not peel or flake off.

A. There are some troubling words here. "Never" scratch or turn color. Even hard anodize can be scratched by some actions. Of course, it will be more scratch resistant than powder coat, so what is "really" meant. The most colorfast black dye will "eventually" turn color if subjected to extreme sunshine or fluorescent lighting. Two step anodizing is an exception as it is inorganic. What you mean and what the plater means has to be put down in writing or you can expect problems. A test run of parts is a great way to find out what is suitable. Oh yes, what does "black" mean to both of you? Hard anodizing of some alloys is a very dark grey--is this black?

A. Since when isn't nickel acetate seal part of MIL-A-8625? The only place that it's nebulous is for Type III hardcoat, however, for Type II - IT IS REQUIRED! There is a fade test within MIL-A-8625 that is long and expensive and only done when specified in the contract, however, there isn't a snowball's chance in hell that an unsealed anodize coating will pass without it. Your anodizing source is just blatantly cutting corners and committing fraud if he's providing unsealed anodize to the military spec and certifying it as such. You should ask your local DCAS inspector for his take (name names!).

A. The problem you describe could be seal related, but more likely it is due to the use of inferior dye. No, they can't be re-dyed. Even if the original seal were imperfect, additional sealing has taken place in situ. If they must be black, and you can't have them stripped and re-anodized, the only solution is to paint them black.

A. All type II anodize is sealed per Mil-Spec. Type III is not.

A. MIL-A-8625F indeed requires sealing of Class 2 (dyed) finishes -- see Paragraph 3.8.1.2. MIL-A-8625F may be downloaded from ASSIST Quick Search: http://assist1.daps.dla.mil/quicksearch/ [Ed. note: that link is now broken, see https://quicksearch.dla.mil/qsSearch.aspx].

A cheaper dye with less fade resistance may have been used. The described parts require disassembly or very time-consuming masking in order to reprocess.

A. Hello,

Okay - 2 things. The first is, a seal IS required for class 2 coatings. Whichever anodizer told you this in the case was mistaken (to be generous). Paragraph 3.8.1.2 clearly states that when class 2 is specified sealing is to be done after dyeing and recommends nickel acetate, boiling water, or sodium dichromate.

Second, there are a few different factors involved in fading, sealing is the most common, but you might also have been given an overly thin coating (preventing sufficient dye absorption) or a poor quality dye might have been used.

Either way, I would be very surprised if further black dye would be absorbed just by immersing the part and then trying to seal it better; odds are you'd get little to no result. If corrosion isn't a factor, you could try other methods to make the part black, such as using an opaque staining color from Dykem. If corrosion is a factor, you're probably best off starting over.

On the flip side... I don't think anything in the dye or seal would hurt the steel parts; so you don't stand to lose much if you go ahead and try it.

Either way, good luck. And, if it was the same anodizer that did the black parts that said sealing wasn't a requirement, I'd REALLY look for a different anodizer.

A. http://www.anodizingindustries.com/Specs/A8625F.pdf [Ed. note: that link is now broken]

There's a link to the current mil spec. If you read it, it does specify that class 2 shall be sealed, although it leaves the sealing medium up to the anodizer. The important thing the mil spec also states is that the coating must pass lightfastness tests, and salt spray tests. Both of these tests should be easily passed if the part is properly sealed. If I were you, in the future, ask your coater to certify that his coating meets that spec. In order to be able certify that coating, he should have sent samples out to an independent lab for testing. Those results should be in his files. I cannot imagine that a quality coater would not seal a type 2, class 2 coating. It's standard practice.

A. Sealing is an often short-cutted end-of-the-line dip.The proper conversion of the aluminum oxide to hydrated aluminum oxide is very important to prevent fading and corrosion. The water must be very pure. Many shops do not understand how easily a seal tank becomes contaminated with drag-in of tap water. Nor do they control the pH and temperature. Nor do they ever perform a "seal test", generic or ASTM. Because the solution ingredients are at such low concentration it is more subject to contamination than anything else in the process line. Many shops do not understand this. They think a weak solution is easy, not so, a weak solution is very critical.

You cannot re-dye your parts. You cannot re-anodize the parts with exposed iron as it will dissolve and fall out and introduce iron as a contaminate into the anodizing solution. Your parts are ruined. You may NOW perform an ASTM seal test and prove that the anodizing was never properly sealed.

Come back to us and let us know the resolution.

A. All Type I and II coatings per MIL-A-8625F must be "completely sealed". Nickel Acetate is not mandated, only a "suitable chemical solution" is required. This specification contains no definition for "completely sealed". One could argue that if your sealing method is sufficient to pass the monthly corrosion resistance test, then is meets the minimum specification requirements. That is unless a light fastness resistance is specified. That has never happened in my 20+ years in the anodizing business. As others have stated, light fastness is dependent on a number of factors other than just sealing.

A. A quick test to see if your parts are sealed is the bleach test. A drop of clorox bleach on the part for about a minute. There should be no discoloration. This is not a substitute for salt spray testing but a quick way to see if it is sealed.

A. Hi Brad. It sounds like, in addition to language and learning barriers, you are concerned about the honesty barrier as well. But in addition to organic dyes and 2-step dyeing, there is also overdyeing a 2-step with an organic, so finding one doesn't mean you don't have the other anyway; then too, poor sealing can lead to fading, as can an anodized layer that is too thin or defective in some other way.

So I think your plan of an accelerated UV test is the most workable answer, but I don't personally know the relationship between X number of hours at Y units of irradiance at wavelength Z vs. real-world fade resistance. Good luck.

Regards,